This project is designed to identify and locate genetic risk factors for sustained cigarette smoking and nicotine addiction. Cigarette smoking remains the largest preventable cause of death in the United States accounting for an estimated 480,000 deaths each year. Unfortunately, one quarter of all Americans continue to smoke, despite the public health warnings and social pressure. Twin studies indicate approximately 50 percent of the variance in smoking initiation and 70 percent of smoking persistence are accounted for by genetic factors. We propose to collect 400 same-sex sibling pairs for use in a genetic linkage study. The probands will be ascertained based on their presentation to a smoking cessation clinic. Both members of the sibling pair will be required to be persistent smokers with at least a 10 year history of smoking and also have a high level of nicotine dependence, as measured by the Fagerstrom Tolerance Questionnaire (FTQ greater than or equal to 7). The siblings will also be evaluated for phenotypes that are known to be associated with smoking. These include a history of major depression and/or alcohol abuse, and certain personality factors. In year 3, 200 sib-pairs and their available parents will be genotyped with 377 microsatellite markers evenly spaced across the human genome. From simulations we have done, we expect to detect one or two of the nicotine susceptibility loci (assuming a three-locus model), but also 3 false-positive signals using a mean test statistic (at p less than 0.001). In year 5, an additional 200 sib-pairs and their available parents will be genotyped with the same markers and we expect that most of the false-positive linkage signals will not be replicated. Simulations indicate that we expect to detect (at p less than 0.0001), on average, 2 genes if there are 2 or 3 loci accounting for the genetic variance and one gene if there are 4 loci. To follow-up these areas of linkage all sibs and available parents will be genotyped with markers spaced at 1-2 cM in the chromosome regions that continue to have the best support for linkage. We also propose to examine the genetic variation in 21 candidate genes from dopaminergic, nicotinic and P450 systems in our sample. Detecting these genes would be a significant first step in our understanding of the genetic variation of addiction to nicotine and possibly other drugs of abuse.